An industrial air compressor waste heat recovery power generation system

Abstract

The application provides an industrial air compressor waste heat recovery power generation system, and relates to the technical field of air compressor waste heat utilization, comprising an air compressor waste heat collection module, a thermoelectric cooperative distribution module, a waste heat recovery heat exchange module, an ORC power generation module, a direct current bus access module and an intelligent control module. The thermoelectric cooperative distribution module continuously distributes the waste heat to the waste heat recovery heat exchange module and the ORC power generation module according to the dynamic distribution instruction generated by the intelligent control module in a variable ratio; the direct current bus access module is provided with a voltage self-adaptive unit, which detects the direct current bus voltage of the air compressor frequency converter in real time and dynamically adjusts the ORC output voltage, and preferentially injects the self-generated power into the direct current bus through a reverse cut-off unit, thereby improving the adaptability of the system to different power air compressors. The improved NSGA-II multi-objective optimization algorithm is combined to solve the optimal thermoelectric distribution ratio in real time, so that the frequent reversing action of the adjusting valve is reduced, the valve jamming risk is effectively reduced, and the energy utilization rate is improved.

Description

Technical Field

[0001] This invention relates to the field of waste heat utilization technology for air compressors, and more particularly to an industrial air compressor waste heat recovery power generation system. Background Technology

[0002] Air compressors are widely used power equipment in industrial production. Their working principle is to convert electrical energy into the pressure energy of compressed air. During the air compression process, a large amount of electrical energy is converted into heat. Only about 10% to 15% of the electrical energy consumed by the air compressor is converted into effective compressed air energy, while the remaining 85% to 90% is converted into heat energy and dissipated in the form of residual heat from lubricating oil, residual heat from compressed air, and heat dissipation from the machine body. Therefore, there is great potential for efficient recovery and utilization of the residual heat of air compressors.

[0003] Early technologies for utilizing waste heat from air compressors involved installing heat exchangers outside the air compressor's lubrication oil or exhaust pipes to transfer waste heat to cold water for hot water production. This method of heat energy utilization was singular and limited by seasonality and heating demand, primarily for power generation. With the development of low-temperature waste heat power generation technology, the Organic Rankine Cycle (ORC) system has been introduced into the field of air compressor waste heat utilization. In the ORC system, an organic working fluid absorbs waste heat from the air compressor in an evaporator and then vaporizes. The high-pressure steam enters an expander to drive a generator, producing electricity. The exhaust steam is liquefied in a condenser and then pressurized and circulated by a working fluid pump, converting low-grade heat energy into high-quality electrical energy, thus overcoming the seasonal limitations of heating scenarios.

[0004] Currently, there are several technical solutions for ORC waste heat power generation. For example, Chinese invention patent application CN120520673A discloses a centrifugal air compressor waste heat power generation system. This system uses high-efficiency heat exchangers between multi-stage centrifugal compressors to transfer the waste heat of compressed air between stages to circulating water. The heat is then transferred to the ORC working fluid via a hot water heat exchanger, driving the working fluid expander to generate electricity. The system is then connected to the plant's power grid via a grid-connected cabinet, achieving cascade recovery of waste heat from multi-stage compression. However, this solution uses AC grid connection at industrial frequency, requiring expensive grid-connected inverter equipment and protection devices. The grid connection approval process is complex, making it difficult to implement quickly in general industrial air compressor stations. The system is only designed for centrifugal air compressors and lacks adaptability to commonly used industrial compressor types such as screw and scroll compressors. Furthermore, it lacks an effective power balance control strategy when power generation fluctuates, affecting grid stability. Chinese invention patent application CN122061860A discloses an ORC DC power replenishment system for interstage waste heat in air compressors. This system addresses the interstage exhaust waste heat (160℃~200℃) in dry oil-free screw air compressors or centrifugal air compressors by constructing a dual-mode thermal diversion mechanism to automatically switch between heat-priority and electricity-priority modes. Its core innovation lies in employing DC clamping direct-drive technology. The electrical energy generated by the ORC power generation module is rectified and boosted before being directly injected into the DC bus of the air compressor's frequency converter. An output voltage higher than the mains rectified voltage is set, and the reverse cutoff characteristic of diodes forces priority consumption of self-generated power, eliminating the need for grid connection and avoiding the costs and approval processes associated with grid-connected equipment. However, this scheme relies on a three-way regulating valve for dual-mode switching, but it does not address the potential jamming problem that may occur when the regulating valve performs high-frequency switching according to proportional current splitting (Reference: "Research Progress on Valve Core Jamming", Luo Yuxuan, Lin Zhenhao, Mu Juan, et al., Mechanical and Electrical Engineering, Vol. 39, No. 7, pp. 865-876, 918, July 2022). If the regulating valve jams, it could very likely lead to the interruption of the entire power generation system. Furthermore, the DC clamping voltage of this scheme is fixed at 550V, failing to consider the differences in DC bus voltage of air compressor frequency converters of different power levels, thus limiting its adaptability.

[0005] Therefore, there is an urgent need for an industrial air compressor waste heat recovery power generation system that can take into account flexible and coordinated heat and power recovery, flexible DC bus adaptation, and intelligent and precise control. Summary of the Invention

[0006] To address the technical problems existing in current air compressor waste heat recovery power generation systems, such as insufficient flexibility in heat and electricity distribution, easy valve core jamming during high-frequency switching of regulating valves, and low DC bus adaptability, this invention provides an industrial air compressor waste heat recovery power generation system. By constructing a heat and electricity coordinated dynamic distribution mechanism, it achieves real-time optimal distribution and intelligent and precise control of air compressor waste heat between heating and power generation.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: An industrial air compressor waste heat recovery power generation system includes an air compressor waste heat acquisition module, a thermoelectric co-distribution module, a waste heat recovery heat exchange module, an ORC power generation module, a DC bus access module, and an intelligent control module that are interconnected. The air compressor waste heat collection module is used to collect the waste heat generated during the operation of the industrial air compressor, including the waste heat of the air compressor lubricating oil or the waste heat of the interstage compressed air. The thermoelectric co-distribution module is connected to the air compressor waste heat acquisition module, receives the waste heat, and, based on the dynamic distribution command output by the intelligent control module, distributes the waste heat to the waste heat recovery heat exchange module and the ORC power generation module in a variable proportion. The waste heat recovery heat exchange module is used to receive a portion of the diverted waste heat, and to prepare hot water through gas-water or oil-water heat exchange and store it in a hot water storage tank; The ORC power generation module is used to receive another portion of the diverted waste heat and drive the ORC cycle to convert thermal energy into electrical energy. The DC bus access module is electrically connected to the DC bus of the ORC power generation module and the air compressor frequency converter, and prioritizes the supply of electrical energy to the main motor of the air compressor. The intelligent control module is communicatively connected to the thermoelectric co-allocation module, the waste heat recovery heat exchange module, the ORC power generation module, and the DC bus access module, respectively. It is used to collect system operating parameters in real time, calculate the optimal thermoelectric allocation ratio based on a multi-objective optimization algorithm, and generate the dynamic allocation command.

[0008] Furthermore, the thermoelectric coordinated distribution module includes: an intelligent proportional regulating valve unit, a flow monitoring unit, and a distribution execution unit; The input end of the intelligent proportional regulating valve unit is connected to the output end of the air compressor waste heat collection module. The intelligent proportional regulating valve unit has a first output end and a second output end. The first output end is connected to the waste heat recovery heat exchange module, and the second output end is connected to the ORC power generation module. The flow monitoring unit is located at the input end and two output ends of the intelligent proportional regulating valve unit, and is used to monitor the total waste heat flow and the flow of each branch in real time. The allocation execution unit is electrically connected to the intelligent proportional regulating valve unit, receives the dynamic allocation command, and drives the intelligent proportional regulating valve unit to adjust the flow allocation ratio between the first output terminal and the second output terminal.

[0009] Furthermore, the intelligent control module includes: a parameter acquisition unit, a demand prediction unit, a multi-objective optimization unit, and an instruction generation unit; The parameter acquisition unit is used to collect system operating parameters in real time. The system operating parameters include: return water temperature, liquid level, and target set temperature of the hot water storage tank; evaporation temperature, evaporation pressure, expander speed, and power generation of the ORC power generation module; air compressor operating load rate, exhaust temperature, lubricating oil temperature, and ambient temperature. The demand forecasting unit predicts the hot water demand trend and waste heat generation trend within a set time window based on the system operating parameters. The multi-objective optimization unit takes the hot water temperature compliance rate and the maximization of power generation as dual optimization objectives, and uses the hot water storage tank capacity constraint, the ORC power generation module rated power constraint, and the intelligent proportional regulating valve adjustment range constraint as boundary conditions to solve the optimal heat and power allocation ratio. The instruction generation unit converts the optimal thermoelectric allocation ratio into the dynamic allocation instruction and sends it to the thermoelectric collaborative allocation module.

[0010] Furthermore, the multi-objective optimization unit uses an improved non-dominated sorting genetic algorithm to solve for the optimal thermoelectric allocation ratio. The improved non-dominated sorting genetic algorithm introduces an adaptive crossover mutation operator and an elite retention strategy on the basis of the traditional NSGA-II algorithm, and designs a penalty function for the constraints of the thermoelectric allocation problem.

[0011] Furthermore, the objective function of the multi-objective optimization unit is to maximize the hot water temperature compliance rate. for:

[0012] in, This represents the actual temperature of the hot water storage tank. Set the target temperature for hot water. For ambient temperature, The heat transfer efficiency coefficient; The objective function of the multi-objective optimization unit is to maximize the power generation. for:

[0013] in, To divert the working fluid mass flow rate to the ORC power generation module, The specific enthalpy of the working fluid at the evaporator outlet. The specific enthalpy of the working fluid at the evaporator inlet. The isentropic efficiency of the expander. For generator efficiency; The boundary conditions include: the liquid level in the hot water storage tank does not exceed the upper safety limit and is not lower than the lower safety limit; the output power of the ORC power generation module does not exceed the rated power; and the opening degree of the intelligent proportional regulating valve is continuously adjustable within the range of 0% to 100%.

[0014] Furthermore, the thermoelectric co-distribution module is also equipped with a dynamic response compensation mechanism. When a sudden change in air compressor load or a step change in hot water demand is detected, the intelligent proportional regulating valve unit first executes a preset emergency switching strategy, and then waits for the multi-objective optimization unit to complete a new round of optimization calculation.

[0015] Furthermore, the ORC power generation module includes: an ORC evaporator, an expander, a generator, a condenser, and a working fluid circulation pump; The heat source side inlet of the ORC evaporator receives waste heat from the second output end of the thermoelectric co-distribution module, and the working fluid side outlet of the ORC evaporator is connected to the working fluid inlet of the expander. The working fluid outlet of the expander is connected to the working fluid inlet of the condenser, and the mechanical output end of the expander is connected to the generator; The working fluid outlet of the condenser is connected to the working fluid side inlet of the ORC evaporator via the working fluid circulation pump, forming a closed-loop circulation of the organic working fluid.

[0016] Furthermore, the DC bus access module includes: a rectifier unit, a DC / DC converter unit, a voltage adaptive unit, and a reverse cutoff unit; The rectifier unit is electrically connected to the generator output terminal of the ORC power generation module, and rectifies the AC power output by the generator into DC power. The DC / DC converter unit is electrically connected to the rectifier unit to perform boost or buck conversion on the DC power. The voltage adaptive unit is used to detect the actual voltage of the DC bus of the target air compressor inverter in real time, and output a voltage setting signal to the DC / DC conversion unit, so that the output voltage of the DC / DC conversion unit adaptively tracks the actual voltage and maintains a preset clamping margin. The reverse cutoff unit is located between the DC / DC conversion unit and the DC bus of the air compressor inverter, and ensures that self-generated power is preferentially injected into the DC bus based on the reverse cutoff characteristics of the diode.

[0017] Furthermore, the intelligent control module also integrates a thermoelectric synergistic evaluation unit, used to calculate the overall thermoelectric efficiency index in real time during system operation:

[0018] in, To effectively generate heat, For effective power generation, This is the electricity conversion factor. This refers to the total amount of waste heat that can be recovered from the air compressor. The thermoelectric synergistic evaluation unit feeds back the thermoelectric integrated efficiency index to the multi-objective optimization unit as prior knowledge for initial population generation and as a basis for dynamic adjustment of constraints.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention utilizes an intelligent proportional control valve unit within a thermoelectric co-distribution module. By introducing an improved NSGA-II multi-objective optimization algorithm, it sets the dual objectives of maximizing hot water temperature compliance and maximizing power generation. It comprehensively considers boundary conditions such as hot water storage tank capacity, ORC rated power, and physical constraints of the control valve to solve for the optimal distribution scheme in real time. This approach avoids the large step movements caused by discrete dual-mode switching in existing technologies, as well as the frequent reversing movements caused by jitter near the threshold. It transforms the valve's adjustment behavior into continuous and smooth micro-displacements, ensuring the stability of system operation.

[0020] This invention uses a voltage adaptive unit in the DC bus access module to detect the DC bus voltage of the target air compressor inverter in real time and dynamically adjust the ORC output voltage, achieving flexible clamping with "one machine, one adapter". This replaces the traditional power frequency AC grid connection scheme, reducing system costs, and is compatible with air compressors of different power levels. Attached Figure Description

[0021] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0022] Figure 1 This is a schematic diagram of the industrial air compressor waste heat recovery power generation system architecture provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a thermoelectric co-distribution module provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an ORC power generation module provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a DC bus access module provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of an intelligent control module provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the working process of an industrial air compressor waste heat recovery power generation system provided in an embodiment of the present invention.

[0023] In the diagram, 01-ORC evaporator heat source side inlet, 02-ORC evaporator working fluid side outlet, 03-expander working fluid inlet, 04-expander mechanical output end, and 05-expander working fluid outlet. Detailed Implementation

[0024] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0025] To overcome the shortcomings of existing air compressor waste heat recovery power generation systems, such as rigid heat and electricity distribution, easy valve jamming, and low DC bus adaptability, this invention provides an industrial air compressor waste heat recovery power generation system. This system achieves real-time optimal allocation and intelligent, precise control of air compressor waste heat between heating and power generation by constructing a dynamic heat and electricity synergy distribution mechanism.

[0026] Figure 1 A schematic diagram of an industrial air compressor waste heat recovery power generation system according to an embodiment of the present invention is shown. Please refer to [link / reference]. Figure 1 As shown, it includes an air compressor waste heat acquisition module, a thermoelectric co-distribution module, a waste heat recovery heat exchange module, an ORC power generation module, a DC bus access module, and an intelligent control module that are interconnected.

[0027] The waste heat collection module for air compressors is used to collect waste heat generated during the operation of industrial air compressors. This waste heat can be waste heat from the air compressor lubricating oil or waste heat from the interstage compressed air. For example, when the air compressor is an oil-injected screw air compressor, the waste heat collection module collects oil-side waste heat through an oil circuit heat exchanger connected in series in the lubricating oil circulation pipeline; when the air compressor is a dry-type oil-free screw air compressor or a centrifugal air compressor, the waste heat collection module collects interstage compressed air waste heat through an air circuit heat exchanger installed on the interstage exhaust pipeline.

[0028] The thermoelectric co-distribution module has its input end connected to the output end of the air compressor waste heat acquisition module, and is used to receive the waste heat. Figure 2 A schematic diagram of a thermoelectric co-distribution module according to an embodiment of the present invention is shown. Please refer to [link / reference]. Figure 2 Specifically, it includes an intelligent proportional control valve unit, a flow monitoring unit, and a distribution execution unit.

[0029] Preferably, the intelligent proportional control valve unit is a linear single-seat control valve driven by an electric actuator, powered by a 24V DC power supply, with a built-in servo motor and position feedback potentiometer, supporting power-off self-locking function. The control signal uses a standard 4-20mA analog current signal as the input command, while simultaneously feeding back a 4-20mA valve position signal to the intelligent control module. The full stroke (0%-100%) response time is ≤5 seconds, and the local response time in the 20-50% commonly used flow division range can be controlled within 2 seconds. Valve position positioning accuracy is ≤±1%, hysteresis ≤0.5%, and dead zone ≤0.3%. The valve body is made of 316L stainless steel, the valve core is overlaid with Stellite hard alloy, and the sealing packing uses pure graphite braided packing with a long-neck valve cover, suitable for high-temperature conditions of 160-200℃. The positioner has a built-in chatter function with a chatter frequency of 80-150Hz and an amplitude of 1-3% of the rated current. A 100-mesh Y-type filter is installed before the valve to intercept any solid particles that may be present in the pipeline.

[0030] The input terminal of the intelligent proportional control valve unit is connected to the output terminal of the air compressor waste heat collection module. The intelligent proportional control valve unit includes a first output terminal and a second output terminal. The first output terminal is connected to the waste heat recovery heat exchange module to supply the heat required for heating; the second output terminal is connected to the ORC power generation module to supply the heat required for power generation.

[0031] In this embodiment, the flow monitoring unit includes three flow measurement points, which are respectively set in the input end pipeline, the first output end pipeline and the second output end pipeline of the intelligent proportional regulating valve unit, for real-time monitoring of the total waste heat flow entering the system and the diversion flow to the heating branch and the power generation branch.

[0032] The input flow measurement point is located between the output end of the air compressor waste heat acquisition module and the input end of the intelligent proportional regulating valve unit. A vortex flow meter is used, which has a temperature range of -40℃ to 350℃ and low pressure loss. During installation, the upstream straight pipe section is ≥10D and the downstream straight pipe section is ≥5D. Flange connection and metal spiral wound gasket sealing are used.

[0033] The first output flow measurement point is located between the first output of the intelligent proportional regulating valve unit and the waste heat recovery heat exchange module. It adopts an electromagnetic flow meter or a clamp-on ultrasonic flow meter. The former has high accuracy (±0.5%) and no moving parts, while the latter does not need to contact the fluid and is suitable for retrofit projects. During installation, the upstream straight pipe section is ≥10D and the downstream straight pipe section is ≥5D.

[0034] The second output flow measurement point is located between the second output of the intelligent proportional control valve unit and the ORC power generation module. It uses a vortex flow meter with a measurement range of 10:1 to 20:1, which can cover a wide range of changes in the power generation branch from low to high flow. The installation requirements are the same as those for the input measurement point.

[0035] All flow measurement points are equipped with PT100 platinum resistance temperature sensors for temperature compensation, and the output signal is a standard 4-20mA analog signal, which is connected to the parameter acquisition unit of the intelligent control module.

[0036] The allocation execution unit is electrically connected to the intelligent proportional regulating valve unit, receives dynamic allocation instructions from the intelligent control module, and drives the intelligent proportional regulating valve unit to adjust the flow allocation ratio between its first output terminal and second output terminal.

[0037] The waste heat recovery heat exchange module is used to receive a portion of the waste heat diverted from the first output end of the thermoelectric co-distribution module. The module contains a gas-water or oil-water heat exchanger, which transfers the high-temperature waste heat to the circulating water to prepare high-temperature hot water, which is then stored in a connected hot water storage tank for use in the plant's production or domestic needs.

[0038] The ORC power generation module is used to receive another portion of waste heat diverted from the second output terminal of the thermoelectric co-distribution module, and convert the thermal energy into electrical energy through the Organic Rankine Cycle (ORC). Figure 3 A schematic diagram of an ORC power generation module according to an embodiment of the present invention is shown, as follows: Figure 3 As shown, the ORC power generation module includes: an ORC evaporator, an expander, a generator, a condenser, and a working fluid circulation pump.

[0039] The heat source side inlet 01 of the ORC evaporator receives waste heat from the second output end, and organic working fluid flows through the working fluid side 02, absorbing heat and vaporizing into high-pressure steam.

[0040] The working fluid inlet 03 of the expander is connected to the working fluid side 02 of the ORC evaporator, and the high-pressure steam drives the expander to rotate and do work.

[0041] The generator is connected to the mechanical output end 04 of the expander, converting the rotational mechanical energy of the expander into electrical energy.

[0042] The condenser is connected to the working fluid outlet 05 of the expander and cools the exhaust steam discharged from the expander into a liquid state.

[0043] The working fluid circulation pump is connected to the condenser and the ORC evaporator working fluid side 02 respectively, and returns the cooled organic working fluid to the ORC evaporator working fluid side 02 to form a closed circulation of the organic working fluid.

[0044] Preferably, the evaporation temperature of the ORC evaporator is preferably in the range of 65-140℃, and the optimal value is dynamically determined according to the heat source temperature.

[0045] For a 90℃ hot water heat source (oil-water heat exchange scenario with an oil-injected screw compressor), the theoretical optimal evaporation temperature is 62℃, and the experimental optimal evaporation temperature is 69℃.

[0046] For a 150°C heat source (after heat exchange of the exhaust gas between oil-free screw stages), the optimal evaporation temperature can reach 139°C.

[0047] The selection principle is based on the requirement that the evaporation temperature should be 5-15℃ lower than the heat source inlet temperature (narrow temperature difference) to balance heat exchange efficiency and heat exchanger economy. Too high an evaporation temperature will increase the heat exchange area and reduce economy; too low an evaporation temperature will reduce the expansion ratio and reduce cycle efficiency.

[0048] The preferred condensing temperature range for the condenser is 25-40°C, with an optimal value of approximately 33°C. The condensing temperature is determined by the temperature of the cooling medium (cooling water or air) and is typically 5-10°C higher than the outlet temperature of the cooling medium. Lowering the condensing temperature can significantly improve circulation efficiency, but it will increase the condenser area.

[0049] In plant areas with sufficient water supply, water-cooled condensation (condensation temperature of approximately 30℃) is preferred, while in water-scarce plant areas, air-cooled condensation (condensation temperature of approximately 40-50℃) is preferred.

[0050] The preferred range of isentropic efficiency of the expander is 65%-85%, with an optimal value of approximately 82.7% (achieved by the vortex expander at an optimal speed of 800 r / min and an optimal pressure ratio of 1.9).

[0051] The isentropic efficiency of screw expanders is typically 60%-75%; The isentropic efficiency of a vortex expander can reach 60%-85%; Run-of-flow turbines can achieve entropy efficiency of up to 85%.

[0052] In this embodiment, a scroll expander or a screw expander is preferably used, and the working fluid circulation pump is a variable frequency canned pump with a volumetric efficiency design value of not less than 60%. The above parameter range is intended to ensure that the system can maintain a high circulating thermoelectric conversion efficiency within the common waste heat temperature range of air compressors.

[0053] Regarding performance under varying operating conditions, considering the frequent load fluctuations of industrial air compressors (typically varying between 30% and 100%), the ORC power generation module possesses wide-load adaptive operation capabilities. When the air compressor is under its calibrated condition at 100% full load, the intelligent proportional control valve unit distributes waste heat according to an optimized ratio, the ORC module evaporation temperature is maintained at the optimal value (e.g., 69℃ or 139℃, depending on the heat source temperature), the scroll expander operates at around 800 r / min, the isentropic efficiency reaches 82.7%, and the system's net power generation efficiency reaches 8%-12% (corresponding to a heat source temperature of 100-150℃).

[0054] When the air compressor load drops to the common operating condition of 60%-80%, the multi-objective optimization unit dynamically increases the proportion of heat flowing to the waste heat recovery heat exchange module. At this time, the ORC module evaporation temperature adaptively decreases by 5-15℃, the expander speed is adjusted accordingly, and the system's net power generation efficiency is 95%-98% of the rated value. Valve operation is smooth, avoiding large abrupt changes. When the air compressor load drops to a low load condition of 30%-50%, the ORC module evaporation temperature is further reduced, and the expander enters a low-speed range. Within this load range, the net power generation efficiency of the ORC system remains at 75%-90% of its rated value. At this time, the dynamic response compensation mechanism is triggered. If the load drops sharply beyond the threshold, the intelligent proportional control valve unit switches to full-load heating mode, and the ORC module enters standby mode; it automatically restarts after the load recovers.

[0055] By adjusting the evaporation temperature setpoint in real time through a multi-objective optimization unit, the system always operates at the optimal operating point under the current load, thereby minimizing the partial load efficiency decay and avoiding the problem of sharp efficiency decay in traditional ORC systems under low load, ensuring the continuity and economy of energy recovery.

[0056] In this embodiment, the circulating working fluid of the ORC power generation module is preferably R245fa (pentafluoropropane) or R1233zd(E) (trans-1-chloro-3,3,3-trifluoropropene). R245fa is chosen because it has excellent thermal stability (critical temperature of about 154°C) and is non-toxic and slightly flammable under operating pressure, which can achieve good temperature matching with the low-grade waste heat of the air compressor at 80°C~100°C. R1233zd(E) is chosen because it has extremely low global warming potential (GWP<7) and zero ozone depletion potential (ODP), with significant environmental protection characteristics. Moreover, its saturated vapor pressure curve is highly consistent with the waste heat temperature range of the air compressor, which can ensure high working efficiency at a higher evaporation temperature.

[0057] The DC bus access module is electrically connected to the DC bus of the ORC power generation module and the air compressor frequency converter. Figure 4 A schematic diagram of a DC bus access module according to an embodiment of the present invention is shown, as follows: Figure 4 As shown, the module includes: a rectifier unit, a DC / DC converter unit, a voltage adaptive unit, and a reverse cutoff unit.

[0058] The rectifier unit is used to rectify the AC power output from the generator into DC power. Preferably, the rectifier unit adopts a three-phase uncontrolled diode full-bridge rectifier circuit, and its output terminal is configured with a filter inductor and capacitor to suppress ripple. In other embodiments of the present invention, the rectifier unit may also adopt a thyristor three-phase fully controlled bridge rectifier circuit to achieve DC-side voltage regulation and soft start by adjusting the firing angle; or a Vienna rectifier or a voltage-type PWM rectifier may be adopted to improve the input-side power factor, reduce harmonics, and enhance adaptability to weak power grids or complex loads.

[0059] The DC / DC converter unit is electrically connected to the rectifier unit and is used to boost or buck DC power so that its output voltage matches the voltage of the DC bus of the target air compressor frequency converter.

[0060] Preferably, the DC / DC converter unit adopts a non-isolated boost topology to achieve boost conversion of the rectified DC power and stably output it to the vicinity of the DC bus voltage of the target air compressor inverter. Specifically, it includes power switching transistors (such as IGBTs or SiC MOSFETs), power diodes (or synchronous rectifier switches), a boost inductor, an output filter capacitor, and drive and control circuitry. Its input terminal is connected to the output terminal of the rectifier unit, and its output terminal is connected to the DC bus of the air compressor inverter via the reverse cutoff unit. To balance efficiency and size, the preferred switching frequency range of the DC / DC converter unit is 20kHz–60kHz, with a conversion efficiency of not less than 95% over the full load range (reaching 96%–98% under rated conditions). The inductance of the boost inductor is designed to ensure stable operation in continuous conduction mode (CCM), with the peak-to-peak output voltage ripple not exceeding 1% of the set output voltage value. The voltage adaptive unit detects the actual voltage of the inverter's DC bus in real time and outputs a voltage setting signal. The output voltage of the DC / DC conversion unit can adaptively track and always maintain a preset clamping margin (such as +5V–+15V) higher than the bus voltage, thereby cooperating with the reverse cutoff unit to achieve priority injection of self-generated power into the DC bus.

[0061] In other embodiments of the invention, a non-isolated Buck-Boost topology may also be used to accommodate a wider range of operating conditions where the input voltage is higher or lower than the bus voltage.

[0062] The voltage adaptive unit is used to detect the actual voltage of the DC bus of the target air compressor inverter in real time, and outputs a voltage setting signal to the DC / DC conversion unit accordingly. This voltage setting signal enables the output voltage of the DC / DC conversion unit to adaptively track the actual bus voltage and maintain a preset clamping margin.

[0063] In this embodiment, the lower limit of the clamping margin is set to 5V. Since the reverse cutoff unit is based on the unidirectional conductivity of a diode, the forward voltage drop of a silicon diode is approximately 0.6-0.7V, and that of a Schottky diode is approximately 0.2-0.4V. To ensure reliable conduction of the diode throughout its entire lifespan, the clamping margin needs to cover the diode's forward voltage drop (0.2-1.6V), the drift of the forward voltage drop with current and temperature (approximately ±0.3V), the DC / DC output voltage sampling error (approximately ±1%), and the bus voltage ripple amplitude (≤5V under normal operating conditions). Setting the lower limit to 5V ensures that even under the most unfavorable operating conditions where the above factors are superimposed, the DC / DC output voltage can still reliably exceed the diode anode conduction requirement, avoiding power injection interruptions caused by the diode's instantaneous cutoff during sudden load changes. The upper limit of the clamping margin is set at 15V. The rated withstand voltage of the inverter's DC bus support capacitor must cover the bus voltage, ripple, and switching spikes, with a margin of 1.2-1.5 times (e.g., for a 540V system, use a film capacitor of 900V or higher). If the clamping margin is too large, the ±5% ripple from the bus itself may approach the capacitor's safe withstand voltage limit or trigger the inverter's overvoltage protection (OU). Setting the upper limit to 15V ensures that in a 540V bus system, the DC / DC output voltage is at most 555V, and after adding the bus ripple, it is in the 528-582V range, far below the safe operating range (720V) of a 900V capacitor, and far from the OU protection threshold. At the same time, the 15V upper limit of the clamping margin can control the boost loss of the DC / DC converter unit and avoid a significant decrease in efficiency.

[0064] The reverse cutoff unit is located between the DC / DC conversion unit and the DC bus of the air compressor frequency converter.

[0065] To overcome the drawbacks of traditional diodes, such as high on-state voltage drop (typically 0.7V-1.5V) under high current, leading to severe heat generation and significant losses, the reverse cutoff unit preferably employs an ideal diode circuit composed of a power switch (such as an IGBT or MOSFET) and its drive control circuit. The power switch is connected in series in the DC circuit, with its body diode orientation aligned with the current-allowed direction. The drive control circuit continuously monitors the voltage polarity across the power switch: when a forward voltage is detected (ORC side voltage is higher than bus side voltage), the drive circuit outputs a low-impedance drive signal to saturate and turn on the power switch. At this time, current flows through the channel instead of the body diode, utilizing the extremely low on-resistance (milliohms) of the power switch to reduce the on-state voltage drop to tens of millivolts, significantly reducing losses and improving system efficiency. When a reverse voltage is detected (bus side voltage is higher than ORC side voltage), the drive circuit quickly turns off the power switch, utilizing its high withstand voltage characteristics to block reverse current flow.

[0066] The intelligent control module is the control core of the entire system, including a parameter acquisition unit, a demand prediction unit, a multi-objective optimization unit, and an instruction generation unit. Figure 5 A schematic diagram of an intelligent control module according to an embodiment of the present invention is shown below. Figure 5 As shown.

[0067] Specifically, the parameter acquisition unit is used to collect the following system operating parameters in real time: return water temperature, liquid level, and target set temperature of the hot water storage tank; evaporation temperature, evaporation pressure, expander speed, and power generation of the ORC power generation module; air compressor operating load rate, exhaust temperature, and lubricating oil temperature; and ambient temperature.

[0068] The demand forecasting unit, based on historical and real-time data acquired by the parameter acquisition unit, uses time series analysis or machine learning algorithms to predict the hot water demand trend and air compressor waste heat generation trend within a set time window (such as the next 15 minutes or 1 hour).

[0069] In this embodiment, the demand forecasting unit has two built-in optional forecasting models, which automatically switch according to the system operation stage and data accumulation. Specifically: In the initial stage of system operation or in scenarios with limited computing resources, the Autoregressive Moving Average (ARIMA) model is adopted. Its input consists of time-series data (sliding window length of 300 sampling points) of hot water tank return water temperature and air compressor load rate, and the output is the point-by-point forecast value for the next 15 minutes. The model order is automatically determined using the Bayesian information criterion, and the parameters are updated online using the recursive least squares method.

[0070] In scenarios with ample historical data and controller acceleration capabilities, an LSTM neural network is employed. The network structure consists of a two-layer LSTM (64+32 hidden units) and a fully connected output layer. The input is a 6-dimensional feature vector (return water temperature, liquid level, load rate, exhaust temperature, ambient temperature, and time encoding), and the output is the predicted total hot water demand and average load rate for the next hour. Training is performed offline pre-training combined with daily online fine-tuning.

[0071] The two models are automatically switched through the model performance monitor: the mean absolute percentage error (MAPE) of the two models is compared, and the model with the smaller error is selected as the current prediction engine. The switching process is smooth and imperceptible.

[0072] The multi-objective optimization unit takes "maximizing the hot water temperature compliance rate" and "maximizing power generation" as dual optimization objectives, and uses the hot water storage tank capacity, ORC power generation module rated power, and intelligent proportional control valve adjustment range as boundary conditions. It uses an improved non-dominated sorting genetic algorithm (Improved NSGA-II) to solve for the optimal heat and power allocation ratio in real time.

[0073] The instruction generation unit converts the optimal thermoelectric distribution ratio solved by the multi-objective optimization unit into specific, executable dynamic distribution instructions (e.g., the intelligent proportional control valve unit should be opened to a certain percentage), and sends them to the distribution execution unit of the thermoelectric collaborative distribution module.

[0074] Figure 6 A schematic diagram of the system workflow provided in an embodiment of the present invention is shown. Please refer to [link / reference]. Figure 6 In this embodiment, the system follows these steps to achieve dynamic thermoelectric synergy during operation: S1, Parameter Acquisition and Trend Prediction: First, the parameter acquisition unit of the intelligent control module starts working, collecting all the aforementioned system operating parameters at a fixed sampling period (e.g., once every 2 seconds). Subsequently, based on this data, the demand prediction unit predicts the hot water demand and waste heat output for a future period and provides the prediction results to the multi-objective optimization unit.

[0075] S2, based on the improved NSGA-II multi-objective optimization solution, uses an improved non-dominated sorting genetic algorithm in the multi-objective optimization unit to solve for the optimal thermoelectric allocation ratio. Details are as follows: S21 performs fitness calculations and constraint checks on every individual in the current population at the beginning of each generation of evolution.

[0076] S211. When the algorithm runs for the first time, an initial population P0 containing N individuals (N=100) is randomly generated. Each individual represents a potential thermoelectric allocation scheme, and its gene is encoded as a real number representing the proportion α (0≤α≤1) of waste heat allocated to the ORC power generation module. Correspondingly, the proportion allocated to the waste heat recovery heat exchange module is 1-α.

[0077] S212, For each individual in the current population, calculate its fitness value for both objective functions: The objective function for maximizing the hot water temperature compliance rate for:

[0078] in, This represents the actual temperature of the hot water storage tank. Set the target temperature for hot water. For ambient temperature, The heat transfer efficiency coefficient; The heat transfer efficiency coefficient The formula used to characterize the actual heat exchange efficiency of a waste heat recovery heat exchange module under current operating conditions is:

[0079] in, The actual outlet water temperature of the waste heat recovery heat exchange module at the current moment (collected in real time by a temperature sensor). The cold water return temperature entering the waste heat recovery heat exchange module; The temperature of the waste heat medium entering the waste heat recovery heat exchange module (i.e., the inlet temperature on the heat source side, such as the air compressor lubricating oil temperature or the interstage exhaust temperature).

[0080] This coefficient reflects the actual efficiency of the heat exchanger in real time. It is effective when there is insufficient water flow, scaling on the heat exchange surface, or fluctuations in the heat source temperature. This will decrease accordingly. The multi-objective optimization unit uses the latest calculated allocation ratio in each round of calculation. Numerical values ​​are substituted into the objective function to ensure that the optimization model always fits the real-time physical state of the system, thus avoiding ineffective optimization caused by the efficiency coefficient deviating from reality.

[0081] Objective function for maximizing power generation for:

[0082] in, To divert the working fluid mass flow rate to the ORC power generation module, The specific enthalpy of the working fluid at the evaporator outlet. The specific enthalpy of the working fluid at the evaporator inlet. The isentropic efficiency of the expander. For generator efficiency; S213, For each individual, check whether it satisfies the following boundary conditions: The liquid level in the hot water storage tank shall not be lower than the lower safety limit (e.g., 20%) and not higher than the upper safety limit (e.g., 95%). The output power of the ORC power generation module shall not exceed its rated power; The opening of the intelligent proportional control valve is continuously adjustable within the range of 0% to 100%.

[0083] For individuals that do not meet the constraints, their fitness value is reduced through a penalty function, causing them to be eliminated during the evolutionary process.

[0084] S22: After the fitness calculation is completed for each generation, an evolutionary iteration and the output of the optimal solution are performed on the entire population to drive the population to evolve toward the optimal solution.

[0085] S221, according to and The value of is used to perform a fast non-dominated sort on all individuals in the current population, dividing them into different Pareto front layers. Simultaneously, the crowding distance between individuals within the same front layer is calculated to maintain population diversity.

[0086] S222 employs a tournament selection method to choose superior parent individuals from the current population. An adaptive crossover operator is then applied to these parent individuals, with a crossover probability P. c Dynamically adjust based on population convergence:

[0087] in, The average crowding distance of the current population. and P represents the maximum and minimum values ​​of the congestion distance, respectively. c The upper limit is 0.9, taken from the typical value recommended by the classic NSGA-II (Deb et al., 2002), to ensure that the solution space is fully explored when the population diversity is at its lowest. c A lower bound of 0.5 is set because when population diversity is at its highest, the crossover probability should not be lower than 0.5 to avoid excessive destruction of superior individuals; otherwise, the algorithm degenerates into a pure mutation search, resulting in excessively slow convergence. The crossover operation uses simulated binary crossover (SBX) with a distribution exponent of 20.

[0088] Then the adaptive mutation operator is executed, with mutation probability P. m In addition to considering population diversity, a temperature deviation factor for hot water storage tanks is also introduced:

[0089] in, This represents the actual temperature of the hot water storage tank. To set the target temperature for hot water, P m The upper limit is 0.1, taken from the typical value recommended by the classic NSGA-II (Deb et al., 2002), to prevent excessive mutation from causing the algorithm to degenerate into a random search. P m With a lower limit of 0.01, the mutation probability should be minimized to maintain stability when population diversity is high and hot water temperature deviation is small. However, at least 1% mutation should be retained to maintain slight population perturbations and avoid getting trapped in local optima. 0.6 and 0.4 are the diversity weighting coefficients. Experimental parameter tuning revealed that in the air compressor waste heat thermoelectric co-allocation problem, population diversity contributes approximately 60% to the algorithm performance, while the temperature deviation factor contributes approximately 40%. This ratio achieved the fastest convergence speed and the smallest Pareto front fluctuation in tests under 10 different operating conditions. The mutation operation uses polynomial mutation (PM) with a distribution exponent of 20.

[0090] Next, a dynamic elite retention strategy is implemented, limiting the number of elite individuals. The calculation formula is as follows: (The formula is dynamically adjusted according to the number of generations g.)

[0091] Where 5 represents the minimum number of elites, 100 represents the population size, and 200 represents the number of iterations.

[0092] At the end of each generation of evolution, the most crowded population is selected from the first non-dominated frontier layer of the merged population R0. One individual is retained as the elite and passed on to the next generation. The remaining individuals are selected based on non-dominance ranking and crowding. This generates the offspring population Q0.

[0093] S223, merge the parent population P0 and the offspring population Q0 into a population R0 of size 2N. Perform non-dominated sorting and crowding calculation on R0 again, and select N optimal individuals to form a new generation population P1.

[0094] Repeat steps S21 and S22 until the preset maximum number of iterations is reached (e.g., 200 iterations in this embodiment). At this point, one solution in the Pareto optimal solution set output by the algorithm is the optimal thermoelectric allocation ratio α_best under the current operating condition.

[0095] It should be noted that S21 is a fundamental computational step that must be executed at the beginning of each generation of evolutionary iteration in S22. Together, they constitute the complete process of the improved NSGA-II algorithm.

[0096] P through diversity perception c Adjustments were made to make the algorithm tend towards conservative search (P) under steady-state conditions (high population diversity). c ≈0.5), the output optimal allocation ratio α_best changes less between adjacent optimization cycles (usually ≤±2%), thereby reducing the number of invalid valve fine-tunings; Increase P when hot water demand changes abruptly. m This enables the algorithm to quickly escape local optima and find new solutions that meet hot water requirements, thereby avoiding large-scale step switching of valves due to optimization lag; The dynamic elite retention strategy retains more elite individuals in the later stages of evolution, making the Pareto front of the algorithm output more stable, the α_best fluctuation in adjacent optimization cycles smaller, and further reducing the frequency of valve adjustment.

[0097] The above design makes the improved NSGA-II algorithm not only a general multi-objective optimization tool, but also a special optimization strategy for the risk of valve jamming in the problem of co-distribution of waste heat and thermoelectricity in air compressors, which is substantially different from the simple transfer of conventional NSGA-II.

[0098] S3, Dynamic Instruction Generation and Execution: The instruction generation unit converts the optimal allocation ratio α_best obtained by the multi-objective optimization unit into a specific dynamic allocation instruction and sends it to the allocation execution unit. The allocation execution unit then drives the intelligent proportional regulating valve unit to precisely adjust its opening to the corresponding position, realizing the proportional distribution of waste heat between heating and power generation.

[0099] It should be noted that although the optimization cycle of this system is 2 seconds, under steady-state conditions, the variation range of the optimal allocation ratio α output by the multi-objective optimization unit is usually maintained within ±2%, corresponding to a valve core displacement of only 0.2-0.5 mm. Such minute displacements have negligible impact on the wear of the valve sealing surface and packing. Therefore, the impact of the system's adjustment strategy on valve life is within an acceptable engineering range.

[0100] S4. Considering the potential for sudden changes in actual operating conditions, the system is also equipped with a dynamic response compensation mechanism. Specifically, when the parameter acquisition unit detects a sudden change in the air compressor load (such as sudden unloading) or a step change in hot water demand (such as a sudden activation of a large amount of hot water from the production line), the conventional optimization calculations of the multi-objective optimization unit may not be able to respond in time. In this case, the intelligent control module will immediately trigger a preset emergency switching strategy. To cover diverse abnormal operating conditions in industrial settings, the system sets four trigger conditions and corresponding emergency strategies, as follows: Triggering condition 1: Sudden change in air compressor load. The parameter acquisition unit detects that the air compressor's operating load rate changes by more than 30% within 1 second, such as suddenly dropping from 100% full load to 0% unloading, or suddenly rising from 0% to 100% load. A sudden load change means that the amount of waste heat generated changes dramatically in an instant. If the original optimized distribution continues, it may cause a certain branch (ORC or heat exchanger) to receive more waste heat than its processing capacity.

[0101] When the load rate drops sharply, the emergency strategy is for the intelligent proportional regulating valve unit to switch to full-capacity heating mode within 0.5 seconds, directing 100% of the waste heat to the waste heat recovery heat exchange module. Since the total amount of waste heat decreases sharply during unloading, priority is given to ensuring hot water supply (the hot water storage tank can act as a buffer) to prevent incomplete vaporization of the working fluid and instability of the expander speed in the ORC evaporator due to insufficient air intake. Therefore, the ORC power generation module suspends receiving waste heat.

[0102] When the load rate suddenly increases, the emergency strategy is for the intelligent proportional regulating valve unit to switch to full power generation mode within 0.5 seconds, that is, to direct 100% of the waste heat to the ORC power generation module. Since the total amount of waste heat increases sharply during loading, the hot water storage tank may already be close to full. To prioritize the use of excess heat for power generation and avoid overheating and overpressure of the hot water, the waste heat recovery heat exchange module suspends receiving waste heat.

[0103] Triggering condition two: A sudden change in hot water demand. The parameter acquisition unit detects that the return water temperature of hot water storage tank 310 drops by more than 15°C within 10 seconds, or the liquid level drops by more than 20% within 10 seconds. A sudden drop in return water temperature or liquid level indicates that a large-scale hot water consuming device has suddenly been started in the plant area (such as a cleaning line, a ruptured water supply pipe in the heating system, etc.), and the demand for hot water increases instantaneously.

[0104] The emergency strategy at this time is for the intelligent proportional regulating valve unit to switch to full-capacity heating mode within 1 second, directing 100% of the waste heat to the waste heat recovery heat exchange module, while the ORC power generation module suspends receiving waste heat. Simultaneously, the intelligent control module sends a standby command to the ORC power generation module, reducing the working fluid circulation pump to its minimum speed to maintain working fluid circulation but not generate electricity. The optimization calculation will restart once the hot water demand returns to normal. Once the return water temperature in the hot water storage tank rises to above 90% of the target set temperature, or the liquid level rises to above 50% of the safety upper limit and remains stable for more than 30 seconds, the system automatically exits the emergency mode and resumes multi-objective optimization calculation.

[0105] Triggering condition three: ORC generator module malfunction, parameter acquisition unit detects any of the following abnormal signals: The evaporation temperature of ORC evaporators exceeds 95% of the critical temperature of the working fluid (for example, the critical temperature of R245fa is 154°C and the trigger threshold is 146°C). The expander's rotational speed exceeds 110% of its rated speed; The generator's output power exceeds 105% of its rated power; The condenser's condensing pressure exceeds 120% of the design pressure; All of these factors prove that the ORC power generation module is malfunctioning, and continuing to supply waste heat to it may cause equipment damage or safety accidents.

[0106] The emergency strategy at this time is for the intelligent proportional regulating valve unit to switch to full-load heating mode within 0.5 seconds, directing 100% of the waste heat to the waste heat recovery heat exchange module, while the ORC power generation module stops receiving waste heat. Simultaneously, the intelligent control module issues an emergency shutdown command to the ORC power generation module, shutting down the working fluid circulation pump and opening the ORC bypass solenoid valve to depressurize the working fluid to the storage tank. The system can only be restarted after the abnormal state of the ORC power generation module is resolved (all parameters return to normal range) and after manual confirmation of reset.

[0107] Triggering condition four: The intelligent proportional control valve unit itself malfunctions, i.e., shows signs of jamming. The intelligent control module detects the following signs of jamming through the valve position feedback signal: The absolute value of the deviation between the command opening and the feedback opening continues to exceed 3% for more than 3 seconds; At the same time, the derivative (rate of change) of the valve position signal approaches zero, meaning that the valve is not actually activated. The valve may have experienced early jamming (slight contamination or increased static friction), and although it is not completely locked, it has deviated from the commanded position.

[0108] The emergency strategy at this time is for the allocation execution unit to immediately execute a self-repairing frequency sweep procedure: within 3 seconds, drive the intelligent proportional control valve unit to slowly move from the current opening to fully open (100%) until the deviation is eliminated, then return to fully closed (0%), and then back to the commanded opening. This frequency sweep action can effectively remove trace contaminants in the gap between the valve core and the valve sleeve, and eliminate static friction.

[0109] If the command-feedback deviation recovers to within ±1% within 5 seconds after the self-repair frequency scan, the self-repair is considered successful and the system returns to normal optimized operation.

[0110] If the command-feedback deviation is not restored after self-repair frequency sweep, or if the stuck symptom is triggered 3 times in a row, it is determined to be a serious stuck situation. The system will switch to fixed ratio heating mode (fixing the intelligent proportional regulating valve unit at the most recent successful opening degree) and issue an alarm signal to notify the operation and maintenance personnel to carry out maintenance.

[0111] It should be noted that the triggering conditions for the emergency switching strategy also include: within a unit time window (e.g., 5 minutes), the number of emergency switching triggers does not exceed a preset threshold (e.g., 3 times). If this threshold is exceeded, the system determines it as a continuous abnormal operating condition, switches to a fixed ratio mode, issues an alarm, and awaits manual intervention.

[0112] The four triggering conditions mentioned above cover the most common abnormal operating conditions in industrial settings. This strategy of responding to emergencies first and then optimizing avoids the valve being driven by high-frequency, reverse commands under sudden changes in operating conditions, thereby effectively reducing the risk of valve jamming.

[0113] In the S5 system, the intelligent control module also integrates a thermoelectric synergistic evaluation unit. This unit calculates the system's overall thermoelectric efficiency in real time. Specifically:

[0114] in, To effectively generate heat, For effective power generation, This is the electrical energy conversion factor (used to convert electrical energy into thermal energy equivalent). This refers to the total amount of waste heat that can be recovered from the air compressor. In this embodiment, according to GB / T 15316-2009 "General Technical Rules for Energy Conservation Monitoring", The value ranges from 2.5 to 3.0. When the system is used to replace electric heating equipment, 3.0 is used; when it is used to replace gas boilers, 2.5 is used. Alternatively, the system can automatically and dynamically calculate the value using the following formula based on local electricity and gas prices. :

[0115] in, The local electricity price (RMB / kWh) The local gas price (yuan / kWh calorific value). The boiler thermal efficiency is set to replace (default 0.92). If no price information is entered, the default compromise value is 2.8.

[0116] This metric is continuously fed back to the multi-objective optimization unit, and it operates throughout the algorithm's iterative optimization process through the following two mechanisms: As prior knowledge for initial population generation: at the beginning of each optimization cycle, the multi-objective optimization unit generates 50% of the initial individuals within a ±10% neighborhood of the optimal allocation ratio α_best obtained in the previous cycle, while the remaining 50% of individuals are randomly generated in the global range [0,1]. This mechanism utilizes the continuity of historical optimal solutions, enabling the algorithm to converge faster under steady-state conditions and reducing unnecessary global searches.

[0117] The basis for dynamic adjustment of constraints: the multi-objective optimization unit is based on... The long-term trend of change is dynamically adjusted to constrain the boundary. (Setting) The benchmark value (Take the average value of the first 100 optimization cycles after the system is put into operation). If the average value of the first 10 optimization cycles is... All below If the level reaches 90%, the lower safety limit for the hot water storage tank level will be raised from 20% to 30%; if this occurs for 10 consecutive optimization cycles... All higher than If the level is 110%, the safe upper limit for the hot water storage tank level will be lowered from 95% to 85%. This mechanism allows the constraints to dynamically adapt to actual operating efficiency, achieving adaptive optimization of system performance.

[0118] Through the two feedback mechanisms mentioned above, the overall thermoelectric efficiency index... It is no longer merely a passive measure of system performance, but has become a driving force that actively guides optimization algorithms toward better solutions, enabling self-learning and continuous improvement of system performance.

[0119] In summary, this embodiment effectively solves the technical problems mentioned in the background art by constructing a complete closed loop of "collection-distribution-utilization-access-control" and focusing on optimizing the continuous thermoelectric ratio adjustment, multi-objective optimization and dynamic response compensation links, thereby realizing the efficient, flexible and intelligent recovery and utilization of air compressor waste heat.

[0120] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. An industrial air compressor waste heat recovery power generation system, comprising an air compressor waste heat acquisition module, a thermoelectric co-distribution module, a waste heat recovery heat exchange module, an ORC power generation module, a DC bus access module, and an intelligent control module, all interconnected; characterized in that, The thermoelectric co-distribution module includes an intelligent proportional regulating valve unit, which is used to receive waste heat and, based on the dynamic distribution command output by the intelligent control module, continuously distribute the waste heat to the waste heat recovery heat exchange module and the ORC power generation module in a variable proportion. The DC bus access module includes a voltage adaptive unit and a reverse cutoff unit; The voltage adaptive unit is used to detect the DC bus voltage of the air compressor frequency converter in real time and dynamically adjust the output voltage of the ORC power generation module so that it adaptively tracks the DC bus voltage. The reverse cutoff unit is used to ensure that spontaneously generated power is preferentially injected into the DC bus; The intelligent control module includes a multi-objective optimization unit, which adopts an improved NSGA-II algorithm. It takes maximizing the hot water temperature compliance rate and maximizing power generation as dual optimization objectives, solves the optimal heat and power allocation ratio in real time, and generates the dynamic allocation command. The improved NSGA-II algorithm introduces an adaptive crossover mutation operator and an elite retention strategy to reduce the adjustment frequency of the intelligent proportional regulating valve unit.

2. The industrial air compressor waste heat recovery power generation system according to claim 1, characterized in that, The thermoelectric coordinated distribution module includes: The intelligent proportional regulating valve unit has its input end connected to the output end of the air compressor waste heat collection module, and has a first output end and a second output end. The first output end is connected to the waste heat recovery heat exchange module, and the second output end is connected to the ORC power generation module. A flow monitoring unit is installed at the input end and two output ends of the intelligent proportional regulating valve unit to monitor the total waste heat flow and the flow of each branch in real time. The distribution execution unit is electrically connected to the intelligent proportional regulating valve unit, receives the dynamic distribution command, and drives the intelligent proportional regulating valve unit to adjust the flow distribution ratio between the first output terminal and the second output terminal.

3. The industrial air compressor waste heat recovery power generation system according to claim 1, characterized in that, The intelligent control module includes: The parameter acquisition unit is used to acquire system operating parameters in real time. The demand forecasting unit, based on the system operating parameters, predicts the hot water demand trend and waste heat generation trend within a future set time window; The multi-objective optimization unit takes the hot water temperature compliance rate and the maximization of power generation as dual optimization objectives, and uses the hot water storage tank capacity constraint, the ORC power generation module rated power constraint, and the intelligent proportional control valve adjustment range constraint as boundary conditions to solve the optimal heat and power allocation ratio. The instruction generation unit converts the optimal thermoelectric allocation ratio into the dynamic allocation instruction and sends it to the thermoelectric coordinated allocation module.

4. The industrial air compressor waste heat recovery power generation system according to claim 3, characterized in that, The multi-objective optimization unit uses an improved non-dominated sorting genetic algorithm to solve for the optimal thermoelectric allocation ratio. The improved non-dominated sorting genetic algorithm introduces an adaptive crossover mutation operator and an elite retention strategy on the basis of the traditional NSGA-II algorithm.

5. The industrial air compressor waste heat recovery power generation system according to claim 3, characterized in that, The objective function of maximizing the hot water temperature compliance rate of the multi-objective optimization unit. for: in, This represents the actual temperature of the hot water storage tank. Set the target temperature for hot water. For ambient temperature, The heat transfer efficiency coefficient; The objective function of the multi-objective optimization unit is to maximize the power generation. for: in, To divert the working fluid mass flow rate to the ORC power generation module, The specific enthalpy of the working fluid at the evaporator outlet. The specific enthalpy of the working fluid at the evaporator inlet. The isentropic efficiency of the expander. For generator efficiency; The boundary conditions include: the liquid level in the hot water storage tank does not exceed the upper safety limit and is not lower than the lower safety limit; the output power of the ORC power generation module does not exceed the rated power; and the opening degree of the intelligent proportional regulating valve is continuously adjustable within the range of 0% to 100%.

6. The industrial air compressor waste heat recovery power generation system according to claim 1, characterized in that, The thermoelectric co-distribution module is also equipped with a dynamic response compensation mechanism. When a sudden change in air compressor load or a step change in hot water demand is detected, the intelligent proportional regulating valve unit first executes a preset emergency switching strategy, and then waits for the multi-objective optimization unit to complete a new round of optimization calculation.

7. The industrial air compressor waste heat recovery power generation system according to claim 1, characterized in that, The ORC power generation module includes: The ORC evaporator receives waste heat from the second output terminal of the thermoelectric co-distribution module at its heat source side inlet. An expander, the working fluid inlet of which is connected to the working fluid side outlet of the ORC evaporator; A generator is connected to the mechanical output end of the expander; A condenser is connected to the working fluid outlet of the expander; The working fluid circulation pump has its inlet connected to the working fluid outlet of the condenser and its outlet connected to the working fluid side inlet of the ORC evaporator, forming a closed-loop circulation of the organic working fluid.

8. The industrial air compressor waste heat recovery power generation system according to claim 1, characterized in that, The DC bus access module includes: The rectifier unit is electrically connected to the generator output terminal of the ORC power generation module to rectify the AC power output by the generator into DC power. The DC / DC converter unit is electrically connected to the rectifier unit to perform boost or buck conversion on DC power. The voltage adaptive unit is used to detect the actual voltage of the DC bus of the target air compressor inverter in real time, and output a voltage setting signal to the DC / DC conversion unit, so that the output voltage of the DC / DC conversion unit adaptively tracks the actual voltage and maintains a preset clamping margin; A reverse cutoff unit is installed between the DC / DC conversion unit and the DC bus of the air compressor inverter. Based on the reverse cutoff characteristic of the diode, it ensures that the self-generated power is preferentially injected into the DC bus.

9. The industrial air compressor waste heat recovery power generation system according to claim 1, characterized in that, The intelligent control module also integrates a thermoelectric synergistic evaluation unit, which is used to calculate the overall thermoelectric efficiency index in real time during system operation. in, To effectively generate heat, For effective power generation, This is the electricity conversion factor. This refers to the total amount of waste heat that can be recovered from the air compressor. The thermoelectric synergistic evaluation unit uses the thermoelectric integrated efficiency index as a priori knowledge or a basis for dynamically adjusting the constraints of the initial population generation in the multi-objective optimization unit.

Images